Re: [3/6] widl: Avoid generating negative offsets for conformant arrays of strings. Take 2.

[Alexandre Julliard]

Dmitry Timoshkov <dmitry(a)baikal.ru> writes:
> This version of the patch limits the changes to conformant arrays of
> strings in order to avoid unwanted changes to other pointer types.

Didn't you say that midl is doing this for all arrays? Why doesn't this
work in our case?

-- 
Alexandre Julliard
julliard(a)winehq.org

[3/6] widl: Avoid generating negative offsets for conformant arrays of strings. Take 2.

[Dmitry Timoshkov]

Alexandre Julliard <julliard(a)winehq.org> wrote:
> > This version of the patch limits the changes to conformant arrays of
> > strings in order to avoid unwanted changes to other pointer types.
>
> Didn't you say that midl is doing this for all arrays? Why doesn't this
> work in our case?

No, midl doesn't do that for all types of arrays, I said that before
I had compared more carefully what widl and midl generate for the whole
dlls/rpcrt4/testss/server.idl. I tried to investigate more deeply what
midl does for all kinds of arrays, but that's enourmous amount of work
(conformant and varying arrays, strings, pointers of different types,
etc.) and I decided to make widl generate working code just for arrays
of strings, that's what I need at this point.

-- 
Dmitry.